jsfs11_-_HighdensityRPMerge-7B-4bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

HighdensityRPMerge-7B - bnb 4bits

• Model creator: https://huggingface.co/jsfs11/
• Original model: https://huggingface.co/jsfs11/HighdensityRPMerge-7B/

Original model description:

tags:

• merge
• mergekit
• lazymergekit
• SanjiWatsuki/Silicon-Maid-7B
• chargoddard/loyal-piano-m7-cdpo
• jsfs11/RandomMergeNoNormWEIGHTED-7B-DARETIES
• NeverSleep/Noromaid-7b-v0.2
• athirdpath/NSFW_DPO_vmgb-7b base_model:
• SanjiWatsuki/Silicon-Maid-7B
• chargoddard/loyal-piano-m7-cdpo
• jsfs11/RandomMergeNoNormWEIGHTED-7B-DARETIES
• NeverSleep/Noromaid-7b-v0.2
• athirdpath/NSFW_DPO_vmgb-7b license: apache-2.0

---

HighdensityRPMerge-7B

HighdensityRPMerge-7B is a merge of the following models using LazyMergekit:

• SanjiWatsuki/Silicon-Maid-7B
• chargoddard/loyal-piano-m7-cdpo
• jsfs11/RandomMergeNoNormWEIGHTED-7B-DARETIES
• NeverSleep/Noromaid-7b-v0.2
• athirdpath/NSFW_DPO_vmgb-7b

🧩 Configuration

models:
  - model: saishf/West-Hermes-7B
    # no parameters necessary for base model
  - model: SanjiWatsuki/Silicon-Maid-7B
    parameters:
      weight: 0.4
      density: 0.8
  - model: chargoddard/loyal-piano-m7-cdpo
    parameters:
      weight: 0.3
      density: 0.8
  - model: jsfs11/RandomMergeNoNormWEIGHTED-7B-DARETIES
    parameters:
      weight: 0.25
      density: 0.45
  - model: NeverSleep/Noromaid-7b-v0.2
    parameters:
      weight: 0.25
      density: 0.4
  - model: athirdpath/NSFW_DPO_vmgb-7b
    parameters:
      weight: 0.2
      density: 0.4
merge_method: dare_ties
base_model: saishf/West-Hermes-7B
parameters:
  int8_mask: true
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "jsfs11/HighdensityRPMerge-7B"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])